Open Access
Add to BibSonomyAbstract
8 mm long channel waveguides were manufactured in polycrystalline diamond with inductively coupled plasma etching and characterized in terms of mode-profile and propagation loss in the 1.5 – 3.4 μm spectral region. As proof of principle, the waveguides were evaluated in an evanescent field sensing setup targeting the CH absorption peak of isopropanol at ~3.4 μm, showing good agreement with numerical finite element simulations.
© 2016 Optical Society of America
1. Introduction
Mid-infrared (mid-IR) spectroscopy is a powerful method to study biomolecules, since it combines chemical specificity and sensitivity with almost non-destructive probing. Mid-IR vibrational spectroscopy (500 to 3500 cm−1; 3-20 µm) is based on the excitation of fundamental molecular vibrations, which are characteristic for each species and molecular structure. Recently, the detection of organic substances with evanescent field absorption in the mid-IR has been demonstrated in slab diamond waveguides (width 100-500 µm, height 14 µm) with performance better than using optimized single- mode GaAs waveguides [1]. Detecting the onset of neurodegenerative diseases before they break out is of high importance for our growing and ageing population, as the societal costs associated with these diseases are presently accelerating rapidly [2,3]. The development of Parkinson’s disease is linked to the different states of α-synuclein [4,5], and it has been shown that there are significant spectral differences in the mid-infrared wavelength range for the different states of this protein [6–8]. We aim to develop a miniaturized mid-IR spectroscopic measurement platform based on diamond waveguides to enable an ultra-sensitive, label-free detection of chemicals and biomolecular interactions. A particular focus would be studies of α-synuclein and correlate quantitative data on appearance of this protein with disease progression to possibly later allow for early diagnosis of Parkinson’s disease and preventive care.
Several platforms transparent in the mid-IR have been proposed and used to demonstrate chemical or biological detection, for instance, silver halide fibers and GaAs waveguides [9], chalcogenide fibers [10] and waveguides [11] as well as Si3N4 waveguides [12] to name only a few. The Si and Ge platforms have also been receiving increased attention [13] due to their great prospects of realizing photonic integrated circuits and low loss waveguides have been extensively characterized for the mid-IR, cf [14–17], with demonstrated sensing in a few cases [18,19].
In this work we have chosen to investigate waveguides in diamond, due to the combination of the advantageous materials properties for mid-IR sensing; transparency, mechanical robustness, relatively low refractive index, biocompatibility and chemical inertness, where the latter is desirable to resist harsh cleaning processes in between measurements. Thanks to novel CMOS compatible processing methods diamond has recently become an important optical material [20]. It has one of the widest transparency window of all optical materials and a low loss far into the mid-IR region (< 1 dB/cm attenuation above 7 µm wavelength) [21,22], which is important for mid IR sensing. Diamond is already a common internal reflectance element in attenuated total internal reflectance – Fourier transform infrared spectroscopy (ATR – FTIR) [23], and it is utilized extensively for bulk Raman lasers [20]. Single-crystal diamond (SCD) waveguides with excellent properties has been reported for several photonics applications in the near-infrared [24–29]. However, SCD is still very expensive and cannot be grown on other substrates than diamond hence the sample size is restricted to < 10x10 mm2. An alternative is polycrystalline diamond (PCD), which is much easier to fabricate and hence is available at a much lower price and in wafer form. However, to date only a few publications have been published on PCD waveguides [1,30]. The first PCD waveguides for the mid-IR were planar and were used to detect acetone in D2O with a limit of detection superior to that of corresponding GaAs waveguides, which indicated the potential of this platform [1].
This paper extends the work presented in [1], now utilizing channel diamond waveguides with a 14x13.5 µm2 cross-section. We describe the fabrication and study the wave guiding properties of these 8 mm long PCD waveguides. The propagation loss and intensity distribution of the guided modes were also characterized in the near- and mid-IR region. Experimental results on evanescent field sensing of simple alcohols show close agreement with finite element numerical simulations. The evanescent field sensing was performed in the lower end of the mid-IR spectrum using two tunable optical parametrical oscillators (OPO) that were swept around the CH-vibrational absorption peak at ~3.4 µm wavelength. To remove end face bevel-angle of ~20 degrees and improve the in/out-put coupling, the waveguides were prepared by focused ion beam (FIB) milling, which also enabled convenient and robust fiber butt-coupling of broadband light into the waveguides.
2. Fabrication of the diamond waveguides
The diamond waveguides (DWG) were etched from a 14 ± 1 µm PCD film on a Si wafer with an optical buffer consisting of a 200 nm nucleation layer of Si3N4 resting on 2 µm SiO2 (Diamond Materials GmbH). An image of the chip and scanning electron microscope (SEM) images of the fabricated waveguides are displayed in Fig. 1. A 4 µm thick Al film was sputtered on the diamond, after which the waveguide geometry was defined in Shipley S1813 photoresist by a standard UV-lithography process. The Al film was etched in BCl3/Cl2 plasma to make a mask for the subsequent diamond etching. The diamond was etched in Ar/O2/SF6 plasma, which produces inclined sidewalls. The sample was then turned over and an Al mask was similarly defined on the backside and the silicon was etched in a standard Bosch process, leaving only a narrow Si frame for the waveguides to rest on. Leftover mask material was removed in piranha solution (hydrogen peroxide in sulphuric acid). For a detailed description of the fabrication procedure see [31]. Smooth vertical end faces on the waveguides, see Fig. 1(c), were produced by Ga ion milling with a focused ion-beam (FIB) system (FEI Strata DB235) for easy input/output coupling by fiber butt-coupling. To enhance the diamond etch-rate and to produce smooth surfaces water vapor was introduced during milling. The finished waveguides had a cross-section of 14x13.5 µm2 and a sidewall bevel-angle of ~20 degrees.
Fig. 1 (a) Image of the diamond waveguide chip where red light is launched through a butt-coupled single mode fiber from the right. The output on the left side is collected with a multimode fiber. (b) SEM image of the waveguides sticking out from the Si-substrate and (c) SEM image of the ion-milled end face of a waveguide. The light grey layer underneath the waveguide is the optical buffer layer consisting of silicon nitride and thermal oxide.
In the SEM image of one of the waveguides, Fig. 1(c), one can see that there is a light gray layer at the bottom of the waveguide, which is the Si3N4/SiO2 optical buffer. This optical buffer layer introduces only a negligible loss of < 0.01 dB/cm for the fundamental guided mode in the wavelength region studied here, i.e., 1.5 – 3.5 µm. The intrinsic attenuation of the PCD is < 0.5 dB/cm for 1.5 – 2 µm wavelength and 5.8 dB/cm at 3.4 µm [22]. Seen from the SEM image, it is also apparent that the sidewalls have a noticeable roughness. This introduces a significant propagation loss, which is discussed in the section below.
3. Experiments and discussion
3.1. Waveguide characterization
Figure 2 displays a schematic of the butt-coupling set-up, which was used to characterize the mode-profiles and the spectral insertion loss (IL) in the 1.5 – 3.5 µm wavelength region. In order to study the basic properties of the DWG a variety of light sources were used: A laser diode (LD) at 1.46 nm wavelength, an amplified distributed feedback (DFB) diode laser at 1.55 µm, a thulium doped fiber laser (TDFL) at 2.05 µm wavelength, and the idler from two different tunable optical parametrical oscillators (OPO) at ~3.4 µm wavelength [32,33]. The LD, DFB, TDFL, and OPOs were polarized before being coupled into single mode fibers (SMF28TM for the LD, the DFB and the TDFL and a ZrF4 fiber for the OPOs) and the input polarization to the waveguide was controlled by means of a fiber polarization controller (PC) before entering the waveguides. At < 2 µm wavelength a standard PC was used whereas for 3.4 µm a rather crude polarization control was achieved by bending the ZrF4 fiber slightly. The output of the DWG was measured either with a charged couple device (CCD) camera, a pyroelectric camera or a power-meter, or the output was butt-coupled to another fiber, which was connected to an optical spectrum analyzer (OSA) or a spectrometer.
Fig. 2 Schematic of the setup used for measuring the insertion loss, mode-profiles and for the evanescent field sensing experiment.
The propagation loss (PGL) for the 8 mm long DWG was calculated by taking into account the total measured attenuation, the coupling efficiency, the propagation length, and the Fresnel reflections. The coupling efficiency between the input fiber and the waveguide was estimated by calculating the overlap of the mode intensity profile between the DWG and the input fiber. The power throughput of the DWG was optimized at each wavelength prior to imaging, after which it was replaced by the fiber. In order to compare the mode field overlap between the fiber and DWG, they were imaged with the same magnification and alignment. The image of the fiber was centered and focused by translating the fiber only, while keeping the camera and objective stationary. Afterward, for each wavelength a two dimensional cross-correlation between the two intensity profiles was performed using the function xcorr2 in MATLAB® [34] in order to optimize the coupling and find the relative position of the fiber and the waveguide. It was important to remove the background noise prior to performing the 2D cross-correlation, otherwise it could result in the wrong relative position. The coupling efficiency was finally calculated as the overlap integral of the electric field distribution of the diamond waveguide, EDWG, with that of the complex conjugate of the fiber, E*Fiber:
Here the electric fields were approximated with the square root of the intensity IDWG and IFiber, i.e., the captured images of the DWG and fiber end-face intensity distribution. The resulting mode field overlap, or coupling efficiency, between the input-fibers and the waveguide was in the range of 60 – 90% depending on the wavelength and the fiber being used.In Fig. 3 the measured PGL is displayed (black squares) for the low loss polarization (transverse electric like mode). In the wavelength region where the diamond is highly transparent, i.e., < 2.5 µm wavelength, the PGL decreases strongly with increasing wavelength, as the primary loss then is coming from side wall scattering [39,40]. The surface roughness of the polished PCD film was measured by interferometry to ~20 nm root-mean-square (rms). Although the sidewall roughness of the DWGs was not measured it can be seen in Fig. 1(c) to be larger than for the top surface. Hence, the sidewall scattering would dominate over surface scattering, which was also confirmed by imaging the waveguides from above with a CCD when guiding light at ~1 µm wavelength. At 3.4 µm wavelength the total scattering loss was estimated to ~6.8 dB/cm by comparing the measured PGL, 12.6 ± 1 dB/cm, with that of the intrinsic attenuation of diamond 5.8 dB/cm [22], cf., the blue dashed curve in Fig. 3.
Fig. 3 The stars in the graph shows the propagation loss as function of wavelength for the diamond waveguides as measured by the individual light sources. Additionally displayed are the intrinsic attenuation of polycrystalline diamond (blue dashed line), mainly caused by multiphonon absorption [22], as well as the simulated propagation loss for the diamond waveguide on insulator (green dash-dotted line).
An additional reason for the waveguide loss could come from focused gallium (Ga) beam micro machining which can cause Ga-contamination of the structured material [35]. Studies has shown that FIB of diamond can give rise to a 4 nm thick gallium oxide layer followed by an almost 50 nm thick gallium-rich amorphous carbon layer [36]. This should be investigated further as there also exist a report showing that FIB of diamond-like films does not affect the optical transmission in the infrared region [37].
To further understand the propagation properties and predict the behavior of these waveguides further up in the mid-infrared region the mode content was simulated for the wavelength region of 2.5 – 10 µm. The simulation was based on the finite element method (FEM) and performed in the Comsol Multiphysics® software, using the mode analysis study of the RF-module, which solves the electromagnetic field distribution in the frequency domain. The waveguide dimensions, as well as the wavelength dependent material properties were specified and the maximum mesh size was less than 1/5 of the local wavelength. The refractive index and the extinction coefficient data of PCD were digitized from [21] and [22], and data for SiO2 and Si3N4 was obtained from [38], respectively. The refractive index and extinction ratio of the surrounding air was set to 1 and 0 respectively. The resulting PGL for the fundamental mode is displayed (in dash-dot green) in Fig. 3 above, neglecting surface/sidewall scattering loss. The blue and the green curve follow each other in the beginning which means that the attenuation induced by the insulator layer is unnoticeable for wavelengths < 5.5 µm. Above 6.5 µm wavelength, on the other hand, the insulator induced loss (green) exceeds the intrinsic PCD attenuation (blue) and increases significantly due to the absorption peak of the thermal oxide at 9.3 µm and silicon nitride at 11.5 µm. However, the material induced PGL for this type of DWG on insulator is calculated to be below 10 dB/cm between 5.7 and 8 µm wavelength, i.e. the region of interest for studies of the amide bands of proteins.
The simulations also shows that with straight sidewall instead of tilted, the weight of the guided mode will shift slightly upwards, i.e., away from the highly absorbing Si3N4/SiO2-layer and the buffer layer induced PGL would decrease by ~4 dB/cm at wavelengths longer than 6.5 µm. Furthermore, having the weight of the mode closer to the sample would of course also be beneficial for the evanescent field sensing.
The simulations were also compared with the measured spatial intensity distribution of the guided modes and there was a close resemblance was found, as displayed in Fig. 4. However, the Abbe diffraction limit of the imaging system was 3.4 µm/(2*NA) = 6.8 µm, where NA is the numerical aperture of the ZnSe microscope objective used, somewhat limited the accuracy in the measurement of the mode-distribution.
Fig. 4 Mode intensity-profile at 3.46 µm wavelength (a) Single-frame excerpt from Visualization 1 showing the mode intensity-profile of the DWG when translating the input ZrF4-fiber 1 µm/frame in the horizontal and vertical direction. (b) Simulated fundamental mode intensity-profile. The images (a) and (b) share the same the color scale, and have been adjusted to match in size.
Visualization 1 is a movie showing the measured, and interpolated, output intensity-profile of the DWG as the input-fiber was translated horizontally and vertically (1 µm/frame). The movie shows that the center of mass of the guided intensity distribution was shifted slightly as the input fiber was translated, however the upper edge of the mode stays constant. This is of course an effect of the physical extent of the waveguide. Nevertheless, the system is essentially robust in terms of transmission versus input fiber position. The transmission dropped only by ~50% when the input fiber was translated ± 7 µm. In the evanescent field sensing measurements presented below, the measured attenuation was hardly affected by changes in the position of the fiber.
3.2. Evanescent wave sensing
As a proof of principle, these DWGs were tested for evanescent field sensing. The setup was the same as shown in Fig. 2 and mid-IR light from the tunable OPOs was sent through the DWG while drops of iso-propanol were deposited on the DWG and the change in attenuation was recorded. The attenuation was measured in two ways and compared with FEM simulations as described above using the refractive index and the extinction coefficient data of iso-propanol from [41].
The first measurement was performed by setting the OPO to specific wavelengths and recording the transmitted power as the iso-propanol was deposited onto the waveguide. The transmission increased by ~20% when the wavelength was off the absorption line (3425 nm, inset Fig. 5, red curve) as the effect of scattering then was reduced by having a lower refractive index contrast between the waveguide and the surrounding. When the OPO wavelength was tuned to the absorption line the transmitted power dropped, as expected. One such measurement is displayed in the inset of Fig. 5 (blue curve) where the power decreased by ~30% when three drops of iso-propanol (0.15 ml) was added, covering the DWG (at t ~10 sec). This temporal measurement was then performed for a few wavelengths around the absorption peak and is plotted (in blue squares) in Fig. 5. By comparing the transmission on and off the absorption line gives and loss of approximately 2.3 dB, or 2.9 dB/cm.
Fig. 5 Measured attenuation as function of wavelength of the DWG with an iso-propanol covering the diamond waveguide. The absorption peak at ~3.37 µm shown by simulation (dashed line) is also visible in the measurements (solid lines). The inset shows the power going through the DWG as a function of time when drops of isopropanol are added onto the waveguide at t = 10 sec.
The second measurement was done by sweeping the wavelength by using either of the two OPOs. The idler spectrum of the continuous wave (CW) OPO [32] was used between ~3320 nm and ~3400 nm and launched through the waveguide. Iso-propanol was kept on top of the DWG while the transmitted power was measured and compared with a reference beam from the OPO. Data was continuously collected while the wavelength of the OPO was slowly tuned. Unfortunately, the power and wavelength retrieval from the OSA, wave meter and power meters only worked at 1-2 Hz, which resulted in that the complete scan took up to 15 min depending on the resolution required. With a faster data acquisition, the wavelength scan would only be limited by how fast the pump of the OPO i.e., the tunable fiber laser, could be scanned, which could be in a few seconds. During the wavelength scan the solution evaporated slightly and was therefore continuously replenished. A reference scan without the iso-propanol was also performed in the same manner. To get an equidistant data set the measured data points were then averaged within a 1 nm window at every 0.5 nm step between 3320 and 3400 nm. The reference scan with the same equidistant data points were then subtracted from the measurement resulting in the green attenuation curve of Fig. 5. Even though the reference beam from the CW OPO was taken from the path going to the mid-IR fiber, it was not perfectly correlated to the power coupled into the fiber, which resulted in a rather noisy curve (3320-3400 nm).
At slightly longer wavelengths, 3400 nm to 3460 nm, the ns pulsed OPO was utilized [33]. The sweep rate was ~1 nm/sec and the measurement with iso-propanol was taken in less than 2 min. The corresponding attenuation curve is shown in Fig. 5 (in red). The peak attenuation is ~2.8 dB/cm here, however as can be seen the noise is higher here.
The absorption peak at ~3370 nm of iso-propanol is clearly visible for both the sequential temporal measurements (in blue squares) and the wavelength swept measurement (in green). The FEM simulations (in dashed turquoise) show a slightly higher attenuation, 3dB/cm, than measured. To partly competing effects can explained this. The first effect is the reduced sidewall scattering loss due to the smaller refractive index step at the interface when the liquid was covering the waveguide. The second effect refers to the multimode nature of the waveguide and the fact that the actual excited mode(s) are not known as described earlier. Higher order modes exhibit greater loss due to larger portion evanescent field and will experience increased attenuation when surrounded by an absorbing material, compared to the simulated first order mode case.
The reduced sidewall surface scattering will shift the attenuation curves in Fig. 5 downwards when adding isopropanol whereas the multimode guidance will shift it upwards. For further studies, with proteins, which might have less absorption than the CH-vibrational peak, it will be paramount to reduce the sidewall roughness and to go towards single mode waveguides to achieve the desired sensitivity, so that the dominant feature affecting the attenuation is primarily the absorption of the studied sample. Decreasing the dimensions of the waveguide will increase the portion of the evanescent field and hence increase the penetration depth into the sample on top. Since we have seen that the sidewall roughness increases with the etch depth in diamond we also expect to see improved sidewall roughness with smaller dimensions. On the other hand, a waveguide with smaller cross-section will also be more sensitive to the surface roughness, choice of buffer and alignment. The sidewall roughness loss can be further reduced by employing rib waveguides [25,42], which, due to their design, not only have less sidewall area but also require even smaller etch depths. On the other hand, a rib waveguide will promote slightly lower portion of evanescent field into the sample since some of it will instead extend into the slab.
There are several other designs that could potentially improve the measurement in terms of sensitivity, either by designing the waveguide cross-sectional design to enhance the evanescent field, e.g., a slot waveguide [43], or by simply to increasing the path length by wrapping the waveguide back and forth over the chip.
4. Conclusions
8 mm long diamond waveguides in poly-crystalline diamond on insulator have been prepared with a 14x13.5 µm2 cross-section and tilted sidewalls. Convenient and robust butt-coupling for broadband input coupling of light was achieved by FIB milling to obtain vertical endfaces. The DWGs were characterized in the wavelength range 1.5 – 3.4 µm and the propagation loss at 3.4 µm was measured to be 12.6 dB/cm, which was attributed to a combination of scattering loss from the tilted sidewalls and the intrinsic material attenuation at this wavelength. A small additional loss contribution from Ga-contamination resulting from end-face FIB cutting should not be excluded. A proof-of-principle experiment by evanescent field sensing of iso-propanol at ~3.4 µm was performed, in which the added attenuation from the iso-propanol on the waveguides was measured. A transmission loss of 2.3 dB (2.9 dB/cm) was obtained when the wavelength was scanned over the absorption line at 3.37µm for isopropanol. This result was in good agreement with numerical finite element simulations that predicted a loss of ~3 dB/cm.
Recent studies [44] have confirmed the usefulness of conventional FTIR-ATR techniques for analyzing protein mixtures and for identifying amyloid fibril information with relevance for Alzheimer’s disease. Next we will try to fabricate single-mode diamond waveguides to further improve the sensitivity of evanescent field IR spectroscopy. Furthermore, several methods exist for surface functionalization of diamond [24] and especially functionalized nanocrystalline diamond (NCD) has been proven to be useful for capturing different types of biomolecules (proteins, DNA etc.) [45–47]. The surface enrichment of biomolecules on the waveguide provided by surface functionalization can thus further increase the sensitivity and specificity. Realizing functionalized single-mode diamond waveguides for protein analysis will be the topic of future studies.
Acknowledgments
This work was supported by grants from VINNOVA through its Uppsala Berzelii Technology Center for Neurodiagnostics and the Swedish Research Council (VR) through its Linnæus Center of Excellence ADOPT, Stiftelsen Olle Engkvist Byggmästare, K.A. Wallenberg Foundation and VR-project 621-2014-5959. Financial support from Magnus Bergvalls Stiftelse, Stiftelsen Tornspiran, Helge Ax:son Johnsons Stiftelse is also acknowledged. Thanks to Peter Zeil, and Cobolt AB, for providing the OPOs [32] and [33], respectively, and to Krzysztof Beć for contributing with the original refractive index and extinction data for iso-propanol [41].
References and links
1. X. Wang, M. Karlsson, P. Forsberg, M. Sieger, F. Nikolajeff, L. Österlund, and B. Mizaikoff, “Diamonds are a spectroscopist’s best friend: thin-film diamond mid-infrared waveguides for advanced chemical sensors/biosensors,” Anal. Chem. 86(16), 8136–8141 (2014). [CrossRef] [PubMed]
2. R. Luengo-Fernandez, J. Leal, and A. Gray, “UK research spend in 2008 and 2012: comparing stroke, cancer, coronary heart disease and dementia,” BMJ Open 5(4), e006648 (2015). [CrossRef] [PubMed]
3. A. Lowin, M. Knapp, and P. McCrone, “Alzheimer’s disease in the UK: comparative evidence on cost of illness and volume of health services research funding,” Int. J. Geriatr. Psychiatry 16(12), 1143–1148 (2001). [CrossRef] [PubMed]
4. M. G. Spillantini, M. L. Schmidt, V. M. Y. Lee, J. Q. Trojanowski, R. Jakes, and M. Goedert, “α-Synuclein in Lewy bodies,” Nature 388(6645), 839–840 (1997). [CrossRef] [PubMed]
5. A. Takeda, M. Mallory, M. Sundsmo, W. Honer, L. Hansen, and E. Masliah, “Abnormal accumulation of NACP/alpha-synuclein in neurodegenerative disorders,” Am. J. Pathol. 152(2), 367–372 (1998). [PubMed]
6. M. S. Celej, R. Sarroukh, E. Goormaghtigh, G. D. Fidelio, J. M. Ruysschaert, and V. Raussens, “Toxic prefibrillar α-synuclein amyloid oligomers adopt a distinctive antiparallel β-sheet structure,” Biochem. J. 443(3), 719–726 (2012). [CrossRef] [PubMed]
7. S. W. Chen, S. Drakulic, E. Deas, M. Ouberai, F. A. Aprile, R. Arranz, S. Ness, C. Roodveldt, T. Guilliams, E. J. De-Genst, D. Klenerman, N. W. Wood, T. P. J. Knowles, C. Alfonso, G. Rivas, A. Y. Abramov, J. M. Valpuesta, C. M. Dobson, and N. Cremades, “Structural characterization of toxic oligomers that are kinetically trapped during α-synuclein fibril formation,” Proc. Natl. Acad. Sci. U.S.A. 112(16), E1994–E2003 (2015). [CrossRef] [PubMed]
8. Y. Cai, C. Lendel, L. Österlund, A. Kasrayan, L. Lannfelt, M. Ingelsson, F. Nikolajeff, M. Karlsson, and J. Bergström, “Changes in secondary structure of α-synuclein during oligomerization induced by reactive aldehydes,” Biochem. Biophys. Res. Commun. 464(1), 336–341 (2015). [CrossRef] [PubMed]
9. B. Mizaikoff, “Waveguide-enhanced mid-infrared chem/bio sensors,” Chem. Soc. Rev. 42(22), 8683–8699 (2013). [CrossRef] [PubMed]
10. P. Houizot, M.-L. Anne, C. Boussard-Plédel, O. Loréal, H. Tariel, J. Lucas, and B. Bureau, “Shaping of Looped Miniaturized Chalcogenide Fiber Sensing Heads for Mid-Infrared Sensing,” Sensors (Basel) 14(10), 17905–17914 (2014). [CrossRef] [PubMed]
11. P. Ma, D.-Y. Choi, Y. Yu, Z. Yang, K. Vu, T. Nguyen, A. Mitchell, B. Luther-Davies, and S. Madden, “High Q factor chalcogenide ring resonators for cavity-enhanced MIR spectroscopic sensing,” Opt. Express 23(15), 19969–19979 (2015). [CrossRef] [PubMed]
12. J. Kasberger, A. Saeed, T. Fromherz, S. Katletz, and B. Jakoby, “Spectral Resolution of the Grating Coupler of a Miniaturized Integrated Evanescent Field IR Absorption Sensor,” IEEE J. Quantum Electron. 47(7), 950–958 (2011). [CrossRef]
13. R. Soref, “Mid-infrared photonics in silicon and germanium,” Nat. Photonics 4(8), 495–497 (2010). [CrossRef]
14. F. Li, S. D. Jackson, C. Grillet, E. Magi, D. Hudson, S. J. Madden, Y. Moghe, C. O’Brien, A. Read, S. G. Duvall, P. Atanackovic, B. J. Eggleton, and D. J. Moss, “Low propagation loss silicon-on-sapphire waveguides for the mid-infrared,” Opt. Express 19(16), 15212–15220 (2011). [CrossRef] [PubMed]
15. Y.-C. Chang, V. Paeder, L. Hvozdara, J.-M. Hartmann, and H. P. Herzig, “Low-loss germanium strip waveguides on silicon for the mid-infrared,” Opt. Lett. 37(14), 2883–2885 (2012). [CrossRef] [PubMed]
16. M. M. Milošević, M. Nedeljkovic, T. M. Ben Masaud, E. Jaberansary, H. M. H. Chong, N. G. Emerson, G. T. Reed, and G. Z. Mashanovich, “Silicon waveguides and devices for the mid-infrared,” Appl. Phys. Lett. 101(12), 121105 (2012). [CrossRef]
17. M. Brun, P. Labeye, G. Grand, J.-M. Hartmann, F. Boulila, M. Carras, and S. Nicoletti, “Low loss SiGe graded index waveguides for mid-IR applications,” Opt. Express 22(1), 508–518 (2014). [CrossRef] [PubMed]
18. P. T. Lin, V. Singh, J. Hu, K. Richardson, J. D. Musgraves, I. Luzinov, J. Hensley, L. C. Kimerling, and A. Agarwal, “Chip-scale Mid-Infrared chemical sensors using air-clad pedestal silicon waveguides,” Lab Chip 13(11), 2161–2166 (2013). [CrossRef] [PubMed]
19. Y. Chen, H. Lin, J. Hu, and M. Li, “Heterogeneously Integrated Silicon Photonics for the Mid-Infrared and Spectroscopic Sensing,” ACS Nano 8(7), 6955–6961 (2014). [CrossRef] [PubMed]
20. R. P. Mildren and J. R. Rabeau, Optical Engineering of Diamond, 1st ed. (Wiley-VCH Verlag GmbH & Co. KGaA, 2013).
21. P. Dore, A. Nucara, D. Cannavò, G. De Marzi, P. Calvani, A. Marcelli, R. S. Sussmann, A. J. Whitehead, C. N. Dodge, A. J. Krehan, and H. J. Peters, “Infrared properties of chemical-vapor deposition polycrystalline diamond windows,” Appl. Opt. 37(24), 5731–5736 (1998). [CrossRef] [PubMed]
22. T. P. Mollart, K. L. Lewis, C. S. J. Pickles, and C. J. H. Wort, “Factors affecting the optical performance of CVD diamond infrared optics,” Semicond. Sci. Technol. 18(3), S117–S124 (2003). [CrossRef]
23. M.-M. Blum and H. John, “Historical perspective and modern applications of Attenuated Total Reflectance-Fourier Transform Infrared Spectroscopy (ATR-FTIR),” Drug Test. Anal. 4(3-4), 298–302 (2012). [CrossRef] [PubMed]
24. C. E. Nebel, B. Rezek, D. Shin, H. Uetsuka, and N. Yang, “Diamond for bio-sensor applications,” J. Phys. D Appl. Phys. 40(20), 6443–6466 (2007). [CrossRef]
25. L. J. McKnight, M. D. Dawson, and S. Calvez, “Diamond Raman waveguide lasers: completely analytical design optimization incorporating scattering losses,” IEEE J. Quantum Electron. 47(8), 1069–1077 (2011). [CrossRef]
26. P. Rath, M. Hirtz, G. Lewes-Malandrakis, D. Brink, C. Nebel, and W. H. P. Pernice, “Diamond nanophotonic circuits functionalized by dip-pen nanolithography,” Adv. Optical Mat. 3(3), 328–335 (2015). [CrossRef]
27. B. J. M. Hausmann, I. Bulu, V. Venkataraman, P. Deotare, and M. Lončar, “Diamond nonlinear photonics,” Nat. Photonics 8(5), 369–374 (2014). [CrossRef]
28. P. Latawiec, V. Venkataraman, M. J. Burek, B. J. M. Hausmann, I. Bulu, and M. Lončar, “On-chip diamond Raman laser,” Optica 2(11), 924–928 (2015). [CrossRef]
29. B. Feigel, H. Thienpont, and N. Vermeulen, “Design of infrared and ultraviolet Raman lasers based on grating-coupled integrated diamond ring resonators,” J. Opt. Soc. Am. B 33(3), B5–B18 (2016). [CrossRef]
30. P. Rath, S. Khasminskaya, C. Nebel, C. Wild, and W. H. P. Pernice, “Diamond-integrated optomechanical circuits,” Nat. Commun. 4, 1690 (2013). [CrossRef] [PubMed]
31. P. Forsberg and M. Karlsson, “Inclined surfaces in diamond: broadband antireflective structures and coupling light through waveguides,” Opt. Express 21(3), 2693–2700 (2013). [CrossRef] [PubMed]
32. P. Zeil, N. Thilmann, V. Pasiskevicius, and F. Laurell, “High-power, single-frequency, continuous-wave optical parametric oscillator employing a variable reflectivity volume Bragg grating,” Opt. Express 22(24), 29907–29913 (2014). [CrossRef] [PubMed]
33. S. Sinisalo and H. Karlsson, “OPO-based photoacoustic spectroscopy speeds up gas analysis,” Photon. Spectra 48, 30–34 (2014).
34. http://se.mathworks.com/help/signal/ref/xcorr2.html
35. X. C. Lehrer, L. Frey, S. Petersen, M. Mizutani, M. Takai, and H. Ryssel, “Defects and gallium contamination during focused ion beam micro machining,” 2000 International Conference on Ion Implantation Technology Proceedings, Ion Implantation Technology 2000 (Cat. No.00EX432) (IEEE), 695–698 (2000). [CrossRef]
36. Y. W. R. McKenzie, M. Z. Quadir, M. H. Gass, and P. R. Munroe, “Focused Ion beam implantation of diamond,” Diamond Related Materials 20(8), 1125–1128 (2011). [CrossRef]
37. Z. Y. Q. Fu, N. K. A. Bryan, and D. Z. Xie, “Investigation of microfabrication of diamond-like film via focused ion beam milling,” Rev. Sci. Instrum. 74(8), 3689–3692 (2003). [CrossRef]
38. http://refractiveindex.info/
39. P. K. Tien, “Light waves in thin films and integrated optics,” Appl. Opt. 10(11), 2395–2413 (1971). [CrossRef] [PubMed]
40. F. P. Payne and J. P. R. Lacey, “A theoretical analysis of scattering loss from planar optical waveguides,” Opt. Quantum Electron. 26(10), 977–986 (1994). [CrossRef]
41. N. Michniewicz, J. Grabska, K. B. Beć, and J. P. Hawranek, “Dielectric functions of iso-propanol and di-iso-propylether in the infrared,” J. Mol. Liq. 203, 143–152 (2015). [CrossRef]
42. Y. Zhang, L. J. McKnight, Z. Tian, S. Calvez, E. Gu, and M. D. Dawson, “Large cross-section edge-coupled diamond waveguides,” Diamond Related Materials 20(4), 564–567 (2011). [CrossRef]
43. C. A. Barrios, “Optical slot-waveguide based biochemical sensors,” Sensors (Basel) 9(6), 4751–4765 (2009). [CrossRef] [PubMed]
44. W. Yang, O. Auciello, J. E. Butler, W. Cai, J. A. Carlisle, J. E. Gerbi, D. M. Gruen, T. Knickerbocker, T. L. Lasseter, J. N. Russell Jr, L. M. Smith, and R. J. Hamers, “DNA-modified nanocrystalline diamond thin-films as stable, biologically active substrates,” Nat. Mater. 1(4), 253–257 (2002). [CrossRef] [PubMed]
45. A. Härtl, E. Schmich, J. A. Garrido, J. Hernando, S. C. R. Catharino, S. Walter, P. Feulner, A. Kromka, D. Steinmüller, and M. Stutzmann, “Protein-modified nanocrystalline diamond thin films for biosensor applications,” Nat. Mater. 3(10), 736–742 (2004). [CrossRef] [PubMed]
46. K. Fromell, P. Forsberg, M. Karlsson, K. Larsson, F. Nikolajeff, and L. Baltzer, “Designed protein binders in combination with nanocrystalline diamond for use in high-sensitivity biosensors,” Anal. Bioanal. Chem. 404(6-7), 1643–1651 (2012). [CrossRef] [PubMed]
47. P. Wang, W. Bohr, M. Otto, K. M. Danzer, and B. Mizaikoff, “Quantifying amyloid fibrils in protein mixtures via infrared attenuated-total-reflection spectroscopy,” Anal. Bioanal. Chem. 407(14), 4015–4021 (2015). [CrossRef] [PubMed]
